Kinase pathway database: an integrated protein-kinase and NLP-based protein-interaction resource

Area of Science:

• Biochemistry
• Bioinformatics
• Systems Biology

Background:

• Protein kinases are vital regulators of cellular functions.
• Understanding protein kinase signaling pathways is crucial for deciphering organism characteristics.
• Existing data on protein kinases are fragmented across various sources.

Purpose of the Study:

• To develop an integrated database for protein kinases from major completely sequenced eukaryotes.
• To provide comprehensive information including classification, functional conservation, orthologs, interactions, and structural data.
• To enable automatic pathway visualization and cross-species comparisons.

Main Methods:

• Development of the Kinase Pathway Database, integrating data from sequenced eukaryotes.
• Utilizing natural language processing (NLP) and a custom dictionary (GENA) for automatic extraction of protein, gene, and compound interactions from abstracts.
• Incorporating phrase patterns for accurate information extraction.

Main Results:

• The database includes protein kinase classification, functional conservation, and ortholog tables.
• It contains over 47,000 protein interactions and protein-gene/compound interaction data.
• An automatic pathway graphic image interface and cross-species comparison tools are provided.

Conclusions:

• The Kinase Pathway Database offers a unified platform for studying protein kinase signaling.
• NLP-driven data extraction enhances the comprehensiveness and accessibility of interaction information.
• The database facilitates comparative pathway analysis across diverse eukaryotic species.